{Minutes} TTWG Meeting 2020-07-16

Thanks all for attending today's TTWG meeting. Minutes can be found in HTML format at https://www.w3.org/2020/07/16-tt-minutes.html


In text format:

   [1]W3C

      [1] https://www.w3.org/


                Timed Text Working Group Teleconference

16 July 2020

   [2]Previous meeting. [3]Agenda. [4]IRC log.

      [2] https://www.w3.org/2020/07/09-tt-minutes.html

      [3] https://github.com/w3c/ttwg/issues/130

      [4] https://www.w3.org/2020/07/16-tt-irc


Attendees

   Present
          Andreas, Atsushi, Gary, Mike, Nigel, Pierre

   Regrets
          Cyril

   Chair
          Gary, Nigel

   Scribe
          nigel

Contents

    1. [5]This meeting
    2. [6]The codecs parameter should have a formal definition of
       the use of the combination operators.
       w3c/tt-profile-registry#71
    3. [7]IMSC 1.2
    4. [8]CSS font-matching algorithm may introduce fingerprinting
       issues w3c/ttml2#1202
    5. [9]text tracks not supported w3c/media-capabilities#157
    6. [10]Meeting close

Meeting minutes

  This meeting

   Nigel: [iterates through agenda]. We have a late AOB about the
   announcement text for
   … IMSC 1.2 Rec.
   … Any more Other Business, or points to make sure we cover?

   group: [no other business]

  The codecs parameter should have a formal definition of the use of the
  combination operators. w3c/tt-profile-registry#71

   github: [11]https://github.com/w3c/tt-profile-registry/issues/

   71

     [11] https://github.com/w3c/tt-profile-registry/issues/71


   Nigel: [summarises issue]
   … I think the semantics are clear but we should check in if we
   agree!
   … The next point is to check where any new text has to go, in
   the registration text or
   … elsewhere in the document.

   Mike: codecs is defined by DASH rfc6831 so having a formally
   defined parameter for TTML
   … is going to confuse people. We need a note that this isn't
   the codecs you think it is, or
   … something like that.

   Nigel: I half remember discussing this before - we're defining
   a parameter for the MIME
   … type, but the DASH codecs is not part of a MIME type.

   Mike: Just a note for now, I will dig around more to see if we
   need any text about this.

   Nigel: Checking in on the semantics,
   … this is about signalling processor requirements
   … and the + operator means "both things on each side of the
   operator" are required.
   … and the | operator means "either thing is acceptable"
   … Then, when both + and | are used, the + has higher
   precedence,
   … so A|B+C|D means any of "A, something that supports B and C,
   or D"
   … and that's it.
   … Do we have agreement on that being the intention?

   group: [no dissent from this]

   Nigel: OK I think we're agreed on that.

   Mike: I have a clarification on rfc6381

   <mike_> RFC6381 defines MIME type parameters "codecs" and
   "profiles" for ISO BMFF-wrapped content

   Mike: I think its okay because we're in the
   application/ttml+xml space but we are going
   … to need a note that this is for the sidecar native filetype
   not ISOBMFF.

   Nigel: Good point, please could you raise the issue on the
   repo?

   Mike: Sure

   Nigel: I think that's orthogonal to defining codecs

   Mike: Agreed

   Nigel: The next question is where we put the text.
   … I'm not sure.

   Mike: It's a parameter so it needs to go in the registration
   text. We're modifying the
   … string and the semantics of the MIME type parameter.

   Nigel: We're not actually modifying it but we are explaining it
   better.

   Mike: If you put the text in the registration part then it
   needs to go to IANA, and if you don't
   … then that's weird.

   Mike: Is this for TTML1 or TTML2 or both?

   Nigel: It's both, we moved the registration text into the
   profile registry.

   Mike: I recall now.

   Nigel: That's really useful guidance. The last question I have
   is if we think we must
   … define these operators completely with reference to TTML2
   profile semantics or if
   … we can so it only partially, e.g. in relation to the any() or
   all() for | and + but with the
   … combination only here in the profile registry.

   Mike: It seems okay to point to the spec for the definition.

   Nigel: I don't want to change TTML2 at this stage.

   Mike: It should really be the same.

   Nigel: I think there's enough wriggle room there.

   Mike: It'd be good if they were the same.

   Nigel: This is enough clarity for me to try to make progress.
   Any other questions or thoughts?

   group: [nothing more]

   SUMMARY: @nigelmegitt to draft a pull request matching the
   above discussion

  IMSC 1.2

   Atsushi: We need to provide some text for a Rec announcement.
   If we want we may
   … coordinate a press release, but we do not need to.
   … We need at least some text to include in a news announcement
   when we go to Rec.

   Nigel: Can I ask to work with you on that announcement text
   Pierre?

   Pierre: Of course, how do you want to do it. Shall I take a
   first pass at it?

   Nigel: Yes please

   Pierre: What is the deadline?

   Atsushi: Release at the Rec. Minimum 2 weeks.

   Pierre: I'll aim to have it ready in the next 2 weeks.

   Atsushi: I need to request publication of Rec with this text
   ready for the internal comms
   … publication team of W3C.

   Pierre: No, where did you post the requested information.

  Nigel: I think the answer to Pierre's question is that it's in
   the agenda issue for this meeting.

   Pierre: I found it. All right, I will work on that.

   Nigel: One other thing, Pierre, thank you for updating the pull
   request, I will take a look at
   … that change to reference the substantive changes document,
   and then we should be good
   … to merge.

   Pierre: That needed to be manually added because it's not part
   of Respec. It was removed
   … from the SoTD when we went back to WD and was never included
   again.

   Nigel: I understand, thank you.

  CSS font-matching algorithm may introduce fingerprinting issues
  w3c/ttml2#1202

   github: [12]https://github.com/w3c/ttml2/issues/1202


     [12] https://github.com/w3c/ttml2/issues/1202


   Nigel: I finally got round to setting up a doodle for this, not
   everyone has been able to
   … respond yet.

   Pierre: Unfortunately I cannot make the two current most likely
   dates. It looks like Sam has the most restricted availability.

   Andreas: I agree with Pierre, Sam's availability is most
   restricted, so maybe we should ask
   … him for some proposed slots in the next two weeks?

   Nigel: Good idea, I will.

   SUMMARY: @nigelmegitt to ask @samuelweiler for additional
   proposed slots.

   Andreas: I wonder if our meeting would be an option too?

   Pierre: Regrets from me for Thursday 23rd July, most likely.
   I'd be available following the meeting.

   Nigel: That's an option I could add.

  text tracks not supported w3c/media-capabilities#157

   Nigel: Mike I think you wanted to bring [13]https://github.com/

   w3c/media-capabilities/issues/157 to our attention?

     [13] https://github.com/w3c/media-capabilities/issues/157


   Mike: CTA Wave is looking at adopting this work on capabilities
   and I just took a quick
   … look at it because it overlaps with work in ATSC and I wanted
   to see where it is going.
   … I noticed they support audio and video but not text tracks.
   … Even if there's only one thing, like the profile, that you
   get back from this API, it struck
   … me as a high level omission. I wanted to know if anyone else
   had any thoughts about this.
   … Exploratory discussion here. I suspect that the reaction will
   be "do a pull request" when you
   … know the list you want to enumerate, or something like that.

   [14]Media Capabilities Editor's Draft

     [14] https://w3c.github.io/media-capabilities/


   Nigel: I think the general intent here is to discover
   processing "grunt" rather than a more
   … general profile discovery mechanism.

   Gary: There's a case to be made that if a processor can decode
   text tracks natively, then
   … why not?
   … I think they may have left it out because for the initial
   work the actual decoding of video
   … has been the focus because you can't really create a polyfill
   for it whereas for text tracks
   … you can. That may be why they have been focusing on just
   media decoding.

   Pierre: Another wild guess is that often timed text is still
   processed in JS by the client.

   Gary: yes

   Nigel: Not what they would consider "hardware" by the platform.

   Mike: So you think the focus is more on the silicon
   capabilities not HTML5?

   Pierre: Yes

   Nigel: +1

   Pierre: Defining "silicon" very loosely there!
   … That's just a guess based on what I've heard in the past.
   … But that's not a reason not to include timed text, but
   probably why it did not cross the
   … minds of the folk working on it.

   Mike: Right, there's effectively a media hierarchy, video ->
   audio -> timed text

   Pierre: I think it's a good idea to raise the issue. There's an
   issue tracker, so maybe we
   … ought to raise an issue.

   Mike: I've done that. John Simmons, who is very active in CTA
   Wave, posted a comment
   … that means "it doesn't matter", saying there are no IMSC
   encoders, which I corrected him on.

   Pierre: When the problem really surfaces, then the document
   will be changed.
   … This document came up because somebody was having trouble,
   say figuring out whether
   … to send HDR content to the platform.
   … Maybe a way to look at it is when there's a problem, the
   document will get modified at that time.

   Nigel: It turns out that Media Capabilities alone isn't
   sufficient for player implementations
   … to decide which media resources to fetch, because for
   instance it doesn't tell you the
   … screen resolution and frame rate (combined) capabilities. So
   something else is needed.

   Mike: Often people find that a profile alone isn't adequate for
   describing what media resources
   … should be fetched. It could be that profile [of TTML] is
   enough on its own.
   … I wanted folk to be aware of this, it may be we can close in
   a year with no action.

   Gary: Looking at the introduction, it wouldn't respond with a
   "can play" response to a
  … MIME type that isn't audio or video. Because text tracks have
   mainly been provided out
   … of band, the methods don't report back ability to play, so
   that's why it may not be there.
   … It would help if media capabilities would strictly declare
   it's about processing capability.

   Mike: It would be [scribe missed]

   Nigel: Think about text track cue timing accuracy too - it
   could be that low end devices
   … have a harder problem with this than high end devices, so it
   may be useful to signal
   … the precision available.

   Mike: That's a good point.

  Meeting close

   Nigel: Thanks everyone for the interesting discussion today.
   Let's finish a couple of minutes
   … early, having completed our agenda. [adjourns meeting]


    Minutes manually created (not a transcript), formatted by
    [15]scribe.perl version 121 (Mon Jun 8 14:50:45 2020 UTC).

     [15] https://w3c.github.io/scribe2/scribedoc.html